Künstliche Intelligenz bei Bildauswertung und Diagnosefindung

Analysis of the role of artificial intelligence in pediatric radiography

“…With the continued development of deep learning and convolutional neural networks, artificial intelligence is rapidly influencing various areas of radiology. However, compared to popular subspecialties in radiology, such as heart imaging, neuro imaging, oncology imaging and breast imaging, paediatric imaging is often neglected [2]. The reason for this may be that the amount of data obtained for adult imaging is much greater than that for paediatric imaging, and the most important thing that AI does need is large training data sets; the more data sets there are, the more accurate the AI will be [2].…”

“…However, compared to popular subspecialties in radiology, such as heart imaging, neuro imaging, oncology imaging and breast imaging, paediatric imaging is often neglected [2]. The reason for this may be that the amount of data obtained for adult imaging is much greater than that for paediatric imaging, and the most important thing that AI does need is large training data sets; the more data sets there are, the more accurate the AI will be [2]. Then each child in paediatric radiology has too much variability in the data due to age differences, while the radiologist may make different assessments based on their own experience and physical fatigue at the time, which makes for discrepancies in the dataset and causes the AI to be less accurate [2].…”

“…The reason for this may be that the amount of data obtained for adult imaging is much greater than that for paediatric imaging, and the most important thing that AI does need is large training data sets; the more data sets there are, the more accurate the AI will be [2]. Then each child in paediatric radiology has too much variability in the data due to age differences, while the radiologist may make different assessments based on their own experience and physical fatigue at the time, which makes for discrepancies in the dataset and causes the AI to be less accurate [2]. Furthermore, since in many cases children may not be as quiet as adults in receiving diagnostic imaging, radiologists are required to take more humane approaches to their examinations, making the data more heterogeneous [2].…”

“…In the field of paediatric medical imaging, traditional medical imaging diagnostic methods may not always be applicable to children due to the differences in their body structure and physiological characteristics from those of adults [1]. At the same time, paediatric diseases have a significant impact on children's health, making accurate and rapid diagnosis of paediatric diseases crucial [2]. The development of artificial intelligence technology has provided new ideas and methods for paediatric medical imaging diagnosis, which can greatly improve the diagnostic accuracy and efficiency of medical imaging by using artificial intelligence algorithms such as deep learning [2].…”

“…At the same time, paediatric diseases have a significant impact on children's health, making accurate and rapid diagnosis of paediatric diseases crucial [2]. The development of artificial intelligence technology has provided new ideas and methods for paediatric medical imaging diagnosis, which can greatly improve the diagnostic accuracy and efficiency of medical imaging by using artificial intelligence algorithms such as deep learning [2]. In addition, there is relatively less research on artificial intelligence in paediatric medical imaging, so there are still more gaps and challenges in the field that need further in-depth research.…”